Introduction

Intracranial arterial occlusion can lead to ischemic necrosis of brain tissue in local blood supply area, and the existence of collateral circulation plays an important role in maintaining cerebral perfusion and metabolism [1]. Patients with good collateral circulation can avoid the occurrence of cerebral infarction and thus have a good prognosis [2]. At present, digital subtraction angiography is the reference in evaluating collateral circulation. The invasive property however limits its clinical applications [3].3D pseudo-continuous arterial-spin-labeling (3D-pcASL) is a well established non-invasive perfusion imaging to quantitatively measure cerebral blood flow (CBF). Different post-label-delay (PLD)s applied in 3D-pcASL allow for evaluating CBF at different flow velocity comprehensively [4]. 3D territorial-ASL (t-ASL) was a further developed ASL technique based on 3D-pcASL, allowing for CBF measurement of specific vessel labeled. With these advantages of both techniques, the formation and the source of collateral circulation on the side of vascular occlusion may be observed [5]. Therefore, this study aimed to evaluate the formation of collateral circulation between symptomatic and asymptomatic patients after severe stenosis or occlusion of unilateral middle cerebral artery (MCA) by using 3D-pcASL at two different PLDs and t-ASL. The prognosis of patients was also assessed.

Materials and methods

Subjects 49 patients (28 male vs 21 female, age: 63.27±12.02 years) with severe stenosis or occlusion of unilateral MCA, defined as the stenotic degree ≥ 75% by MRA, were included in this study. These patients were classified into symptomatic group (n=26) and asymptomatic group (n=23) according to whether notable symptoms occurred in the blood supply area of diseased vessels for half a month. Symptomatic group included patients with cerebral infarction and patients with transient ischemic attack (TIA).
MRI experiment 3D-pcASL and t-ASL measurements were performed on a 3T-MRI scanner (GE signa pioneer, USA) with 29-channel head & neck coil for each patient. For 3D-pcASL，PLDs of 1.5 and 2.5ms were applied. For t-ASL, bilateral internal carotid arteries, ipsilateral external carotid artery and vertebrobasilar artery were separately labelled. Other scan parameters were listed in Table 1. Total scan time was about 30 mins.
Image analysis 3D-pcASL and t-ASL data were post-processed in vendor-provided software on GE Advanced Workstation 4.6. 3D-pcASL derived CBF and vessel-specific CBF mapping from t-ASL were obtained for each patient accordingly.On CBF maps of 3D-pcASL, regions of interest (ROI)s with size of 15-20cm2 were manually drawn on the blood supply region of MCA close to lateral ventricle. PLD specific CBF levels of both affected and contralateral MCA were obtained for each patient. Vessel-specific CBF maps from t-ASL were used to assess the formation of collateral compensation in the blood supply area of affected MCA between two patient groups. Together with MRA, t-ASL was also used to analyze the opening of secondary collateral circulation.
Statistical Analysis All statistical analyses were performed in SPSS 22.0. Independent sample T-test or Mann-Whitney U-test was used for measurement data, and Chi-square test was applied for count data. PLD specific CBF values of MCA supply area at two patient groups were used for inter-group comparison with independent sample T-test, and intra-group comparison with paired sample T-test. Receiver-operating-characteristic (ROC) curve analysis and area-under-the-curve (AUC) were used to evaluate the separate diagnostic efficacy of CBF at PLD1.5s and PLD2.5s in distinguishing symptomatic from asymptomatic patients. Chi-square test or Fisher's test was used to compare the collateral compensation of the blood supply area of affected MCA on vessel-specific CBF maps between two patient groups. The sources of collateral compensations were counted. P<0.05 was considered statistical significance.

Results

Comparable clinical factors, including age, gender, hypertension, diabetes mellitus, smoking and drinking, were found between two patient groups (Table.2). Compared with asymptomatic group, symptomatic group showed significantly lower CBF in both affected and contralateral side of MCA at both PLDs (all P<0.05;Table.3).CBFs at both PLDs were significantly different between the affected and contralateral side for symptomatic group (P<0.05). Meanwhile, this difference was only revealed at PLD1.5s for asymptomatic group (P <0.05). Additionally, on the affected side, the CBF increase of MCA from PLD1.5s to 2.5s was statistically higher in asymptomatic than symptomatic group (P<0.001). Using ROC analysis, CBF at PLD2.5s showed higher clinical efficacy than at PLD1.5s in distinguishing symptomatic from asymptomatic patients (Figure.1).On vessel-specific CBF maps, compared with asymptomatic group, symptomatic group showed significantly less collateral compensation in the affected MCA (P<0.05; Figs2,3). Meanwhile, significantly higher patients with secondary collateral compensation were found in asymptomatic than symptomatic group (P<0.05). Moreover, for patients with collateral compensation, the compensations were mostly found from vertebrae-basilar artery, and the rest were from the contralateral internal carotid artery or ipsilateral external carotid artery.

Discussion and Conclusion

In this study, relative to symptomatic patients, asymptomatic patients revealed stronger collateral perfusion, and larger number of patients with secondary collateral compensatory opening. This might explain why asymptomatic patients had no symptoms, and symptomatic patients showed different outcomes depending on the amount of collateral compensation. In conclusion, combined multi-PLD 3D-pcASL and t-ASL might provide a comprehensive assessment of collateral compensation for patients with unilateral middle cerebral artery severe stenosis or occlusion.

Acknowledgements

NONE

References

[1] Wei W, Yi XY, Ruan JH, et al. Influence of collateral circulation on cerebral blood flow and frontal lobe cognitive function in patients with severe internal carotid artery stenosis[J]. BMC Neurol, 2019, 19(1):151.[2] Hartkamp NS, Petersen ET, Chappell MA, et al. Relationship between haemodynamic impairment and collateral blood flow in carotid artery disease[J]. J Cereb Blood Flow Metab, 2018, 38(11):2021-2032. [3] Verdolotti T, Pilato F, Cottonaro S, et al. ColorViz, a New and Rapid Tool for Assessing Collateral Circulation during Stroke[J]. Brain Sci, 2020, 10(11):882. [4] Choi HJ, Sohn CH, You SH, et al. Can Arterial Spin-Labeling with Multiple Postlabeling Delays Predict Cerebrovascular Reserve ?[J]. AJNR Am J Neuroradiol, 2018, 39(1):84-90. [5] Hartkamp NS, Petersen ET, De Vis JB, Bokkers RP, Hendrikse J. Mapping of cerebral perfusion territories using territorial arterial spin labeling: techniques and clinical application[J]. NMR Biomed, 2013, 26(8):901-12.